Stem Cell-derived Human Cardiomyocytes (SC-hCMs) offer great potential for improving the accuracy of pre- clinical cardiac safety testing. We have characterized a population of SC-hCMs and have demonstrated that these cells show sensitive pharmacology that accurately predicts clinical responses. However, due to low assay throughput and limited resources, only 15 reference compounds were tested. We now propose to increase throughput in preclinical electrophysiology (EP) screens by utilizing a higher-throughput automated EP instrument. We will expand our preclinical in vitro testing to 77 compounds that have been carefully selected based on their known torsadogenic and/or QT prolonging effects. Results from SC-hCM-based assays will be referenced against complete (8 concentration, 8 replicate) concentration-curves of the same compounds generated from high-throughput screens of cell lines expressing each of the major cardiac ion channels. A statistics-based model will be created in collaboration with the U.S. Food and Drug Administration and a leading in silico modeling firm, Leadscope. This model will be based on the unique databases we create which, together the development of software dedicated to mining public and proprietary cardiac databases will dramatically increase productivity of pre-clinical cardiac safet screening. The set of services and products that will result from this project have the potential t save millions of dollars annually by reducing attrition of marketed but cardiotoxic drugs, to improve the safety of drugs in development, and to increase efficiency of drug development by allowing companies to focus on the most promising and safe drug candidates.